Dear STL-cutters,
As I understand, the problem concentrates on one point:
The software and/or milling controller does not have the ability to
fully understand when to interpolate a curved milling path towards a
small bending radius of the surface, or when to stop and mill a sharp
corner.
Sure, you can:
1. Do a smooth interpolation. But then the one or the other small
feature may get wiped away.
2. Follow the polygon closely. But this will result in coarse surfaces
and problems with the spindel dynamics, stumbling from one triangle to
the other.
Result:
For the more complicated cases of life, you have to decide manually, as
described below. The info what you MEAN with the data is just not there.
"Edge or non-edge" is the question.
Regards
Juergen
Juergen Bauer, Siemens AG, EC CS A PD
E-Mail: Juergen.Bauer@spy.siemens.de
Adress: Siemensstr. 13, 67346 Speyer, Germany
Phone: +49-6232-30-2501; Fax -2110
http://w2.siemens.de/ec/ecb/connect/auto/auto_b.htm
----Original Message-----
From: Matt Michaelis [SMTP:michma@carpediem.com]
Sent: Monday, October 19, 1998 10:02 PM
To: SCat3D@aol.com; rp-ml@bart.lpt.fi
Subject: Re: machining from .stl files
Steve C,
I did not mean to challenge or question your expertise, or knowledge of
point
cloud data. I was trying to better understand what you have
accomplished, and,
in doing so, best understand how it can be utilized by understanding its
limitations as well as its strengths.
Unfortunately, one major point I left out of my previous post, was my
desire to
automatically or more quickly create a surface file from point cloud
data. Back
to the original question, when going from an STL file to CNC via nurb
surfaces
(since my tooling department tells be that nurb surfaces are much, much,
much
better for CNC than STL files) a quick, accurate conversion is very
desirable. I
can and have accomplished accurate nurb surface creation from 3D point
cloud data
and STL files, but it is currently a laborious task using my methods
(Pro/E and
Surfacer). Therefore, any method or software that speeds this process
up is of
great interest to me.
Bottom line: does your process/software accomplish nurb surface
creation
faster? If so, I am very interested whatever specifics/examples you are
willing
to share. I am often in need of a source for auto-surfacing or a
software
package that can accomplish it.
Flat surfaces in an STL file definitely make my job much easier also.
However, I
was trying to understand how or if, provided your methods are
quick/automatic,
you were able to deal with large flat section in an STL file.
I am not convinced that true auto-surfacing is possible using NURBS,
however, I
am convinced that there are or will be better ways than my current
methods.
Maybe your method/software is.
Respectfully,
B. Matthew Michaelis - Product Development Engineer
Compression, Inc.
25242 Arctic Ocean Drive
Lake Forest, CA 92630
(949)586-5875
http://www.carpediem.com
Matt.Michaelis@carpediem.com
SCat3D@aol.com wrote:
> <<As a side note, creating a nurb surface model from 3D scan data
(data from a
> 3D
> scanner) is a different animal than creating one from an STL file.>>
>
> Matthew,
>
> As a reformed 3D Systems employee with a decade of experience in RP
processes,
> I am well aware of the differences between uniform scan data and a CAD
> generated file. As a software coder, I am fully capable of creating
solutions
> to such problems. Contrary to your opinion, large flat areas actually
make
> the job easier, as they can act as easily replicable datums.
>
> The software we use is not currently available to the public. I have
tried
> the other programs and I am aware of their capabilities and
limitations.
> Bottom line: we are DOING it.
For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/
For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/
This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:46:56 EEST